whether support Fast-dLLM v2

[FWXT]

Fast-dLLM v2 has below Generation Process to speed up:

1. Block-level Generation: Autoregressive at the block level
2. Sub-block Parallelization: Parallel decoding within blocks for efficiency
3. Hierarchical Caching: Block and sub-block level caching for speed optimization
   whether already support it? thx!

[drewjin]

We are supporting Fast-dLLM v2

[drewjin]

#7

[drewjin]

@FWXT Already supported